2021年最新最全Flink系列教程__Flink高级特性和新特性(八)

package cn.itcast.flink.broadcast;

import org.apache.flink.api.common.state.BroadcastState;
import org.apache.flink.api.common.state.MapStateDescriptor;
import org.apache.flink.api.common.state.ReadOnlyBroadcastState;
import org.apache.flink.api.common.typeinfo.Types;
import org.apache.flink.api.java.tuple.Tuple;
import org.apache.flink.api.java.tuple.Tuple2;
import org.apache.flink.api.java.tuple.Tuple4;
import org.apache.flink.api.java.tuple.Tuple6;
import org.apache.flink.configuration.Configuration;
import org.apache.flink.streaming.api.datastream.BroadcastStream;
import org.apache.flink.streaming.api.datastream.DataStreamSource;
import org.apache.flink.streaming.api.datastream.SingleOutputStreamOperator;
import org.apache.flink.streaming.api.environment.StreamExecutionEnvironment;
import org.apache.flink.streaming.api.functions.co.BroadcastProcessFunction;
import org.apache.flink.streaming.api.functions.source.RichSourceFunction;
import org.apache.flink.streaming.api.functions.source.SourceFunction;
import org.apache.flink.util.Collector;

import java.sql.Connection;
import java.sql.DriverManager;
import java.sql.PreparedStatement;
import java.sql.ResultSet;
import java.text.SimpleDateFormat;
import java.util.Date;
import java.util.HashMap;
import java.util.Map;
import java.util.Random;

/**
 * Author itcast
 * Date 2021/6/24 8:29
 * 两个数据流  1.事件流 2.用户配置流 3.connect关联操作 4.打印输出 5.执行任务
 * 
 * {"userID": "user_3", "eventTime": "2019-08-17 12:19:47", "eventType": "browse", "productID": 1}
 * {"userID": "user_2", "eventTime": "2019-08-17 12:19:48", "eventType": "click", "productID": 1}
 * 
 * 'user_2', '李四', 20
 * 最终的数据流  6个 Tuple6
 * (user_3,2019-08-17 12:19:47,browse,1,王五,33)
 * (user_2,2019-08-17 12:19:48,click,1,李四,20)
 */
public class BroadcastStateDemo {
    public static void main(String[] args) throws Exception {
        //1.env
        StreamExecutionEnvironment env = StreamExecutionEnvironment.getExecutionEnvironment();
        //设置并行度
        env.setParallelism(1);
        //2.source
        //-1.构建实时数据事件流-自定义随机
        //
        DataStreamSource<Tuple4<String, String, String, Integer>> clickSource = env.addSource(new MySource());
        //-2.构建配置流-从MySQL
        //<用户id,<姓名,年龄>>
        DataStreamSource<Map<String, Tuple2<String, Integer>>> configSource = env.addSource(new MySQLSource());
        //3.transformation
        //-1.定义状态描述器
        //MapStateDescriptor>> descriptor =
        //new MapStateDescriptor<>("config",Types.VOID, Types.MAP(Types.STRING, Types.TUPLE(Types.STRING, Types.INT)));
        MapStateDescriptor<Void, Map<String, Tuple2<String,Integer>>> broadcastDesc = new MapStateDescriptor<>("config", Types.VOID, Types.MAP(Types.STRING, Types.TUPLE(Types.STRING, Types.INT)));
        //-2.广播配置流
        //BroadcastStream>> broadcastDS = configDS.broadcast(descriptor);
        BroadcastStream<Map<String, Tuple2<String, Integer>>> broadcastDS = configSource.broadcast(broadcastDesc);
        //-3.将事件流和广播流进行连接
        //BroadcastConnectedStream, Map>> connectDS =eventDS.connect(broadcastDS);
        SingleOutputStreamOperator<Tuple6<String, String, String, Integer, String, Integer>> result = clickSource.connect(broadcastDS)
                //-4.处理连接后的流-根据配置流补全事件流中的用户的信息
                .process(new BroadcastProcessFunction<Tuple4<String, String, String, Integer>, Map<String, Tuple2<String, Integer>>, Tuple6<String, String, String, Integer, String, Integer>>() {

                    @Override
                    public void processElement(Tuple4<String, String, String, Integer> value, ReadOnlyContext ctx, Collector<Tuple6<String, String, String, Integer, String, Integer>> out) throws Exception {
                        //读取出来 f0 为 userId
                        //事件流中读取用户 userId
                        String userId = value.f0;
                        //从ctx环境变量中通过 desc 读取出来广播状态
                        ReadOnlyBroadcastState<Void, Map<String, Tuple2<String, Integer>>> broadcastState = ctx.getBroadcastState(broadcastDesc);
                        //如果广播状态不为空，get(null) 获取出来 配置数据Tuple2
                        if (broadcastState != null) {
                            Map<String, Tuple2<String, Integer>> map = broadcastState.get(null);
                            //判断 map 不为空则
                            if (map != null) {
                                Tuple2<String, Integer> stringIntegerTuple2 = map.get(userId);
                                //取出姓名和年龄
                                //collect 收集 Tuple6
                                //3-4.处理(process)连接后的流-根据配置流补全事件流中的用户的信息,Tuple4和Tuple2合并
                                //处理每一条元素，processElement
                                out.collect(Tuple6.of(
                                        userId,
                                        value.f1,
                                        value.f2,
                                        value.f3,
                                        stringIntegerTuple2.f0,
                                        stringIntegerTuple2.f1
                                ));
                            }
                        }
                    }

                    //value就是MySQLSource中每隔一段时间获取到的最新的map数据
                    //先根据状态描述器获取历史的广播状态 ctx.getBroadcastState(desc)
                    @Override
                    public void processBroadcastElement(Map<String, Tuple2<String, Integer>> value, Context ctx, Collector<Tuple6<String, String, String, Integer, String, Integer>> out) throws Exception {
                        //再清空历史状态 broadcastState 数据
                        BroadcastState<Void, Map<String, Tuple2<String, Integer>>> broadcastState = ctx.getBroadcastState(broadcastDesc);
                        //最后将最新的广播流数据放到 state 中（更新状态数据） broadcastState.put(null,value)
                        broadcastState.clear();
                        broadcastState.put(null, value);

                    }
                });

        //处理广播中的元素
        //4.sinks
        result.print();
        //5.execute
        env.execute();
    }

    /**
     * 
     */
    public static class MySource implements SourceFunction<Tuple4<String, String, String, Integer>> {
        private boolean isRunning = true;
        @Override
        public void run(SourceContext<Tuple4<String, String, String, Integer>> ctx) throws Exception {
            Random random = new Random();
            SimpleDateFormat df = new SimpleDateFormat("yyyy-MM-dd HH:mm:ss");
            while (isRunning){
                int id = random.nextInt(4) + 1;
                String user_id = "user_" + id;
                String eventTime = df.format(new Date());
                String eventType = "type_" + random.nextInt(3);
                int productId = random.nextInt(4);
                ctx.collect(Tuple4.of(user_id,eventTime,eventType,productId));
                Thread.sleep(500);
            }
        }

        @Override
        public void cancel() {
            isRunning = false;
        }
    }

    /**
     * <用户id,<姓名,年龄>>
     */
    public static class MySQLSource extends RichSourceFunction<Map<String, Tuple2<String, Integer>>> {
        private boolean flag = true;
        private Connection conn = null;
        private PreparedStatement ps = null;
        private ResultSet rs = null;

        @Override
        public void open(Configuration parameters) throws Exception {
            conn = DriverManager.getConnection("jdbc:mysql://node3:3306/bigdata?useSSL=false",
                    "root",
                    "123456");
            String sql = "select `userID`, `userName`, `userAge` from `user_info`";
            ps = conn.prepareStatement(sql);
        }
        @Override
        public void run(SourceContext<Map<String, Tuple2<String, Integer>>> ctx) throws Exception {
            while (flag){
                Map<String, Tuple2<String, Integer>> map = new HashMap<>();
                ResultSet rs = ps.executeQuery();
                while (rs.next()){
                    String userID = rs.getString("userID");
                    String userName = rs.getString("userName");
                    int userAge = rs.getInt("userAge");
                    //Map>
                    map.put(userID, Tuple2.of(userName,userAge));
                }
                ctx.collect(map);
                Thread.sleep(5000);//每隔5s更新一下用户的配置信息!
            }
        }
        @Override
        public void cancel() {
            flag = false;
        }
        @Override
        public void close() throws Exception {
            if (conn != null) conn.close();
            if (ps != null) ps.close();
            if (rs != null) rs.close();
        }
    }
}

package cn.itcast.flink.broadcast;

import org.apache.flink.api.common.serialization.SimpleStringEncoder;
import org.apache.flink.api.common.time.Time;
import org.apache.flink.core.fs.Path;
import org.apache.flink.runtime.state.filesystem.FsStateBackend;
import org.apache.flink.streaming.api.datastream.DataStreamSource;
import org.apache.flink.streaming.api.environment.StreamExecutionEnvironment;
import org.apache.flink.streaming.api.functions.sink.filesystem.OutputFileConfig;
import org.apache.flink.streaming.api.functions.sink.filesystem.StreamingFileSink;
import org.apache.flink.streaming.api.functions.sink.filesystem.bucketassigners.DateTimeBucketAssigner;
import org.apache.flink.streaming.api.functions.sink.filesystem.rollingpolicies.DefaultRollingPolicy;

/**
 * Author itcast
 * Date 2021/6/24 10:52
 * Desc TODO
 */
public class StreamingFileSinkDemo {
    public static void main(String[] args) throws Exception {
        //1．初始化流计算运行环境
        StreamExecutionEnvironment env = StreamExecutionEnvironment.getExecutionEnvironment();
        env.setParallelism(1);
        //2．设置Checkpoint（10s）周期性启动 和 stateBackend 存储路径
        // Sink保证仅一次语义使用 checkpoint 和 二段提交
        env.enableCheckpointing(10000);
        env.setStateBackend(new FsStateBackend("file:///d:/chk/"));

        //4．接入socket数据源，获取数据
        DataStreamSource<String> source = env.socketTextStream("node1", 9999);
        //5．创建Streamingfilesink对象
        OutputFileConfig config = OutputFileConfig
                .builder()
                .withPartPrefix("crm")
                .withPartSuffix(".txt")
                .build();
        //5-1. 创建输出文件配置，指定输出路径 /FlinkStreamFileSink/parquet
        StreamingFileSink sink = StreamingFileSink.forRowFormat(new Path("hdfs://node1:8020/FlinkStreamFileSink/parquet"),
                new SimpleStringEncoder<String>("UTF-8"))
                // sink-kafka new FlinkKafkaProducer
                //5-2.StreamingFileSink 行格式化 ， withBucketAssigner->DateTimeBucketAssigner
                .withBucketAssigner(new DateTimeBucketAssigner())
                //withRollingPolicy -> 默认滚筒策略
                .withRollingPolicy(DefaultRollingPolicy.builder()
                        .withMaxPartSize(128 * 1024 * 1024)
                        .withRolloverInterval(Time.seconds(2).toMilliseconds())
                        .withInactivityInterval(Time.seconds(2).toMilliseconds())
                        .build())
        //withOutputFileConfig -> 输出文件的配置
                .withOutputFileConfig(config)
                .build();
        //6.设置输出 sink
        source.print();
        source.addSink(sink).setParallelism(1);
        //7．执行任务
        env.execute();
    }
}

Sink 行格式化 ， withBucketAssigner->DateTimeBucketAssigner
 .withBucketAssigner(new DateTimeBucketAssigner())
 //withRollingPolicy -> 默认滚筒策略
 .withRollingPolicy(DefaultRollingPolicy.builder()
 .withMaxPartSize(128 * 1024 * 1024)
 .withRolloverInterval(Time.seconds(2).toMilliseconds())
 .withInactivityInterval(Time.seconds(2).toMilliseconds())
 .build())
 //withOutputFileConfig -> 输出文件的配置
 .withOutputFileConfig(config)
 .build();
 //6.设置输出 sink
 source.print();
 source.addSink(sink).setParallelism(1);
 //7．执行任务
 env.execute();
 }
 }